JP4714934B2 - Photogrammetry method and photogrammetry program - Google Patents

Description

The present invention relates to a photogrammetry method and a photogrammetry program for obtaining the position coordinates of each part of a rectangular parallelepiped building and the inclination angle of a roof from one piece of two-dimensional image data obtained by photography.

In recent years, research and development of photovoltaic power generation systems have been actively conducted in order to cope with environmental problems such as an increase in energy demand and global warming.
A typical solar power generation system has multiple solar panels arranged as a solar cell array on the roof of a building that is exposed to sunlight. However, the amount of power generated depends on the position of the sun and the surrounding buildings. It is greatly influenced by things.

  Therefore, in order to predict the power generation amount of such a solar power generation system, the latitude, longitude, direction of the place where the system is installed, and the case where the solar cell panel is installed on an inclined roof Requires information such as the inclination angle.

Conventionally, the coordinates of the position of the building where the photovoltaic power generation system is installed and the data of the inclination angle of the roof are different from the method of reading from the drawing of the building, for example, as described in Patent Document 1 It was obtained from a method of calculating based on a plurality of photographs (stereo photographs) taken from.
JP2002-31527

  As for the method of reading from the drawing, there is no means to confirm that the building is expanded or renovated after the drawing is created, and in the method of calculating from the stereo photograph, clear pixel information taken from a plurality of positions is obtained. In order to handle it, there is a problem that enormous amount of image data is required and complicated calculation processing is required.

Therefore, the present invention pays attention to the fact that most buildings have a rectangular parallelepiped outline, and the photo can easily calculate the position coordinates of each part of the building and the inclination angle of the roof from the image data of one photo. The purpose is to provide a surveying method and a photogrammetry program.

The photogrammetry method of the present invention provided for the above-mentioned purpose is to perform edge extraction processing on a single photographic image taken from a position where two side surfaces of a rectangular parallelepiped portion of a building can be seen, and to construct a wire frame of the building A second step of displaying the wire frame on a computer monitor screen, and a coordinate value of each vertex of a reference cuboid set in a virtual spatial coordinate system on the monitor screen. A coordinate value is converted into a coordinate value on a perspective coordinate system having two vanishing points on the left and right sides set virtually, and a perspective view of the reference rectangular parallelepiped is displayed on the monitor screen in combination with the wire frame of the building. 3, an operation of rotating the direction of the reference cuboid with respect to the spatial coordinate system, an operation of expanding / contracting the length in the three directions of width, height, and depth, and both the left and right sides of the perspective coordinate system A fourth step of causing each ridge line of the reference rectangular parallelepiped to coincide with a rectangular parallelepiped portion of the wire frame of the building displayed on the monitor screen by performing an operation of moving the positions of two vanishing points, respectively. The spatial coordinates of the rectangular parallelepiped portion of the building according to the coordinate values of the vertices of the reference rectangular parallelepiped relative to the spatial coordinate system when the cuboid portion of the wireframe of the building and the ridges of the reference cube coincide on the monitor screen A fifth step of determining a position in the system, and display on the monitor screen by rotating the upper surface about one of the ridgelines around the upper surface of the reference cuboid with respect to the spatial coordinate system. A sixth step of aligning the upper surface of the reference rectangular parallelepiped with the roof portion of the wireframe of the building that is being used, and the space coordinate system It is made of a seventh step of determining the inclination angle of the roof of the building from the inclination angle of the quasi-rectangular top surface.

  Also, the photogrammetry program of the present invention causes a computer to perform edge extraction processing on a single photographic image taken from a position where two side surfaces of a rectangular parallelepiped portion of a building can be seen, and create a wire frame of the building A first step; a second step of displaying the wire frame on a monitor screen of the computer; and a coordinate value of each vertex of a reference cuboid set in a virtual spatial coordinate system stored in the memory of the computer. Read from the memory and perform coordinate conversion to a coordinate value on a perspective coordinate system having two left and right vanishing points virtually set on the monitor screen, and a perspective view of the reference rectangular parallelepiped on the monitor screen In response to the operation of the operation unit of the computer, the third step of combining and displaying the wire frame on the spatial coordinate system The direction of the reference cuboid is rotated, the length in three directions of width, height, and depth is expanded and contracted, and the positions of the two vanishing points on both the left and right sides of the perspective coordinate system are moved and displayed on the monitor screen. A fourth step of matching each ridgeline of the reference cuboid to a rectangular parallelepiped part of the wireframe of the building, and a cuboid part of the wireframe of the building and each ridgeline of the reference cube on the monitor screen A coordinate value of each vertex of the reference rectangular parallelepiped relative to the spatial coordinate system is stored in the memory, and a position of the rectangular parallelepiped portion of the building in the spatial coordinate system is determined by the coordinate values; and The building displayed on the monitor screen by rotating the upper surface around one of the ridgelines around the upper surface of the reference cuboid in response to the operation of the operation unit A spatial coordinate system when the upper surface of the reference rectangular parallelepiped coincides with the roof portion of the wire frame of the building on the monitor screen in a sixth step of aligning the upper surface of the reference rectangular parallelepiped with the roof portion of the wire frame; The inclination angle of the upper surface of the reference rectangular parallelepiped is stored in the memory, and the seventh step of determining the inclination angle of the roof of the building in the spatial coordinate system based on the inclination angle is executed.

  According to the present invention, the position coordinates of each part of the building and the angle of the roof can be easily calculated based on one photo image, so that the position coordinates of each part of the building are calculated from the stereo photo image, There is no need to take photos from multiple locations.

  In addition, since the present invention processes the actual image data by a geometric calculation method, a plurality of pieces of pixel data are processed with respect to an enormous number of pixels as in the conventional stereo photo image processing. Since a process for performing a comparison operation is not required, noise is hardly generated, and the position coordinates of each part of the building and the inclination angle of the roof can be easily obtained with high accuracy.

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic configuration diagram of a photogrammetry system for carrying out the present invention. In this embodiment, a photogrammetry system 1 includes a CPU 3 and a memory serving as a storage unit connected to each other by a bus line 2. 4. An external storage device 5 such as a hard disk, an input interface 6, an output interface 7, a keyboard 8 and a mouse 9 as an operation unit connected to the input interface 6, a monitor 10 and a printer 11 connected to the output interface 7, etc. It is configured using a general-purpose personal computer.

  The input interface 6 can be connected to devices such as a digital camera 12, a scanner 13, and a video 14 for capturing photographic image data.

  In addition to the OS (Operating System) for operating this system, the external storage device 5 stores a photogrammetry program, and the CPU 3 reads the photogrammetry program from the external storage device 5 to the memory 4 and executes it. Thus, the personal computer is operated as the photogrammetry system 1.

  In FIG. 1, the memory 4 is composed of a RAM (Random Access Memory). Although not shown, the bus line 2 is also connected to a ROM (Read Only Memory) storing a BIOS or the like.

Next, a procedure of processing performed by the photogrammetry system 1 will be described. The building photographic image data handled by the photogrammetry system 1 is an image captured by the digital camera 12 or an image captured by the optical camera with the scanner 13 or digital image data captured from the video 14 and has a wall surface. The target is a rectangular parallelepiped building.

  The photographic image data is taken from the position where the two sides of the rectangular parallelepiped portion can be seen, and is taken into the photogrammetry system 1 via the input interface 6 and stored in the external storage device 5. Keep it.

FIG. 2 shows a processing flow performed by the photogrammetry system 1. When the photogrammetry program installed in the external storage device 5 of the photogrammetry system 1 is started, the photogrammetry system 1 is saved in the external storage device 5. The photograph image data being read is read into the memory 4 and edge extraction processing is executed to create a wire frame of the photograph image. (Step S1)

  In the present embodiment, the photographic image read into the memory 4 is first displayed on the monitor 10 via the output interface 7, and an instruction to start edge extraction processing from the keyboard 8 or the mouse 9 as the operation unit in the present invention. Upon input, the photogrammetry program executes edge extraction processing of photographic image data, and displays the outline of the building photograph on the monitor 10 as a wire frame. (Step S2)

  FIG. 3 is a wire frame of a building having a general gable roof that is displayed on the screen of the monitor 10 and is displayed as a two-point perspective view. , B, C, F, G, H, I, J, the vanishing point D1 exists on the extension line of the line segment AF and the line segment BC, and the vanishing point D2 exists on the extension line of the line segment FH and the line segment CG. Exists. The positions of these vanishing points D1 and D2 vary depending on the size of the building and the distance between the building and the viewpoint (camera position).

  In the photogrammetry system 1 of the present embodiment, by operating the keyboard 8 or the mouse 9 as an operation unit, a part or the whole of the wire frame is synthesized with the image of the building photograph and displayed on the screen of the monitor 10. You can choose to display only the wireframe. FIG. 4 shows a state in which the image of the building photograph and the wire frame of only the rectangular parallelepiped portion are combined and displayed on the screen of the monitor 10.

  Next, the photogrammetry program sets a virtual reference cuboid in the memory 4 and displays the reference cuboid on the screen of the monitor 10 in combination with a wire frame as a perspective view by a two-point perspective method. (Step S3)

  As shown in FIG. 5, this standard rectangular parallelepiped has initial shapes a, b, c, d, e, f, g, h in a Cartesian coordinate system (x, y, z) which is one of the spatial coordinate systems. The coordinate value of each vertex is stored in the memory 4 as a cube as a vertex.

  FIG. 6 shows a state in which the reference cuboid is transformed from a Cartesian coordinate system to a perspective coordinate system and displayed on the screen of the monitor 10 as a perspective view. In this perspective view, the ridge line ad of the reference cuboid is shown. And vanishing point d1 exists at the intersection point on the extension line of ridge line bc, and vanishing point d2 exists at the intersection point on the extension line of ridge line dh and ridge line cg.

  In this way, the reference rectangular parallelepiped displayed on the screen of the monitor 10 as a perspective view is displayed on the coordinate axes (x, y) of the Cartesian coordinate system as shown in FIG. , z) an operation of rotating the direction around, as shown in FIG. 8, the reference rectangular parallelepiped a, b, c, d, e, f, g, h of the initial shape is converted into the reference rectangular parallelepiped a ′, b ′, c ′, You can perform operations to expand and contract in the direction of each coordinate axis (x, y, z) in the Cartesian coordinate system on d ', e', f ', g', h '. The surveying program updates the coordinate values of the vertices in the Cartesian coordinate system of the reference cuboid stored in the memory 4, and converts these coordinate values into coordinate values in the perspective coordinate system and displays them on the screen of the monitor 10. Display as a perspective view.

  The reference cuboid displayed as a perspective view on the screen of the monitor 10 can change the positions d1 and d2 of the vanishing points on the left and right sides to positions d1 ′ and d2 ′ by operating the operation unit as shown in FIG. With this operation, the correspondence between the Cartesian coordinate system and the perspective coordinate system can be changed.

  By operating the operation unit as described above, rotation of the reference rectangular parallelepiped direction so as to coincide with the line segment of the rectangular parallelepiped portion of the wire frame of the building photographic image data displayed on the screen of the monitor 10, In addition to expanding and contracting, the inclination of the upper and lower ridge lines of the reference cuboid is adjusted according to the upper and lower ridge lines of the rectangular parallelepiped portion of the wire frame of the building on the display screen of the monitor.

  By such an operation, the left and right vanishing points d1 ′ and d2 ′ of the reference rectangular parallelepiped displayed as the perspective view coincide with the left and right vanishing points D1 and D2 of the wire frame image of the building, and are set on the screen of the monitor 10. A virtual perspective coordinate system is specified, and the coordinate values of each vertex in the Cartesian coordinate system of the rectangular parallelepiped part of the building match each vertex a, b, c, d, e, f of the reference rectangular parallelepiped. , g and h are determined as coordinate values. (Step S5)

  Next, the inclination angle of the roof of the building is obtained. This operation is performed from the operation unit, as shown in FIG. 10, on the screen of the monitor 10, the upper surfaces a ′, d ′, e ′ of the reference rectangular parallelepiped matched with the wire frame image of the rectangular parallelepiped portion of the building. , h ′ are rotated by an angle θ about the ridges a ′, d ′ to coincide with the roof surface displayed as a wire frame as inclined surfaces a ′, d ′, i ′, j ′. (Step S6)

  Thus, the inclination angle θ of the roof is determined by calculating the inclination angle of the upper surface when the upper surface of the reference rectangular parallelepiped coincides with the roof surface in the Cartesian coordinate system on the screen of the monitor 10. (Step S7)

Next, FIG. 11 is a figure which shows the building of the roof of the approaching ridge displayed by the two-point perspective method.
In the case of the roof having such a shape, as shown in FIG. 12, by operating the operation unit, the upper surface of the reference rectangular parallelepiped is rotated upward around the ridge line a′d ′ on the screen of the monitor 10, and the line segment i , j is set as k, and the auxiliary line a'k and the auxiliary line d'k of the alternate long and short dash line are displayed, and the line segment ij is adjusted to approach and separate in parallel to the ridge line a'd '. The alternate long and short dash lines a′k and d′ k are made to coincide with the ridge lines on both sides of the wire frame of the roof displayed by combining with the reference rectangular parallelepiped on the screen of the monitor 10.

  By such an operation, even in the case of the roof of the approaching ridge, the inclination angle of each portion of the roof in the Cartesian coordinate system can be determined in the same manner as in the case of the gable roof.

  Also, as shown in FIG. 13, in a building having a combination of a plurality of ridges, a wire frame is displayed for each ridge as shown in FIG. By performing the operation in the same manner, the arrangement of each part of the building and the inclination of each part of the roof can be determined.

It is a schematic block diagram of the photogrammetry system in one Embodiment of this invention. It is a figure which shows the processing flow of the photogrammetry system in one Embodiment of this invention. It is a figure which shows the wire frame displayed like the perspective view drawn by the two-point perspective method on the monitor screen. It is a figure which shows the state by which the photographic image of the building and the rectangular parallelepiped part of the wire frame were combined and displayed on the monitor screen. It is a figure which shows the reference | standard cuboid displayed in the Cartesian coordinate system. It is a figure which shows the reference hexahedron drawn by the two-point perspective method. It is a figure which shows rotation of the reference hexahedron in the space displayed by the two-point perspective method. It is a figure which shows the expansion-contraction deformation | transformation of the reference hexahedron in the space displayed by the two-point perspective method. It is a figure explaining the operation which makes the vanishing point of a reference hexahedron displayed on a wire frame by the two-point perspective method coincide with the vanishing point of a building. It is a figure explaining operation which makes the upper surface of a reference hexahedron match the inclination of the roof of the building displayed with the wire frame, and calculates the inclination angle of the roof. It is a figure which shows the building of the roof of the near building displayed by the two-point perspective method. It is a figure explaining the method of calculating | requiring the inclination of the roof in the case of the roof of a group building. It is a perspective view of the complex-shaped building displayed by the two-point perspective method. It is the figure which displayed the building of FIG. 13 by the wire frame.

  The present invention can be used for predicting the amount of power generated by a photovoltaic power generation system installed in a building, and can also be applied to predicting sunlight conditions behind a building and the direction of wind passing between buildings. Can do.

It is a schematic block diagram in one Embodiment of the photogrammetry system of this invention. It is a figure which shows the processing flow of the photogrammetry system of this invention. It is a figure which shows the wire frame displayed like the perspective view drawn by the two-point perspective method on the monitor screen. It is a figure which shows the state by which the photographic image of the building and the rectangular parallelepiped part of the wire frame were combined and displayed on the monitor screen. It is a figure which shows the reference | standard cuboid displayed in the Cartesian coordinate system. It is a figure which shows the reference hexahedron drawn by the two-point perspective method. It is a figure which shows rotation of the reference hexahedron in the space displayed by the two-point perspective method. It is a figure which shows the expansion-contraction deformation | transformation of the reference hexahedron in the space displayed by the two-point perspective method. It is a figure explaining the operation which makes the vanishing point of a reference hexahedron displayed on a wire frame by the two-point perspective method coincide with the vanishing point of a building. It is a figure explaining operation which makes the upper surface of a reference hexahedron match the inclination of the roof of the building displayed with the wire frame, and calculates the inclination angle of the roof. It is a figure which shows the building of the roof of the near building displayed by the two-point perspective method. It is a figure explaining the method of calculating | requiring the inclination of the roof in the case of the roof of a group building. It is a perspective view of the complex-shaped building displayed by the two-point perspective method. It is the figure which displayed the building of FIG. 13 by the wire frame.

Explanation of symbols

1 Photogrammetry system 2 Bus line 3 CPU (Central processing unit)
4 Memory 5 External storage device 6 Input interface 7 Output interface 8 Keyboard 9 Mouse 10 Monitor 11 Printer 12 Digital camera 13 Scanner 14 Video

Claims (2)

A first step of performing edge extraction processing on a single photographic image taken from a position where two side faces of a rectangular parallelepiped portion of the building can be seen to create a wireframe of the building;
A second step of displaying the wire frame on a computer monitor screen;
The coordinate value of each vertex of the reference cuboid set in the virtual spatial coordinate system is coordinate-converted into a coordinate value on the perspective coordinate system having two vanishing points on the monitor screen virtually set on the monitor screen, A third step of displaying a perspective view of the reference cuboid combined with the wire frame of the building on the monitor screen;
An operation for rotating the direction of the reference cuboid with respect to the spatial coordinate system, an operation for expanding / contracting the length in the three directions of width, height, and depth, and the positions of the two vanishing points on the left and right sides of the perspective coordinate system A fourth step of matching each ridge line of the reference rectangular parallelepiped with the rectangular parallelepiped portion of the wire frame of the building displayed on the monitor screen by performing an operation of moving
On the monitor screen, the spatial coordinate system of the rectangular parallelepiped portion of the building is determined by the coordinate value of each vertex of the reference rectangular parallelepiped relative to the spatial coordinate system when the rectangular parallelepiped portion of the wireframe of the building coincides with each ridgeline of the reference cube. A fifth step of determining a position at
By rotating the upper surface about one of the ridge lines around the upper surface of the reference cuboid with respect to the spatial coordinate system, the roof portion of the wire frame of the building displayed on the monitor screen A sixth step of matching the upper surfaces of the reference cuboids;
A photogrammetry method comprising a seventh step of determining an inclination angle of the roof of the building from an inclination angle of the upper surface of the reference rectangular parallelepiped with respect to the spatial coordinate system. A first step of causing a computer to perform edge extraction processing on a single photo image taken from a position where two side surfaces of a rectangular parallelepiped portion of the building can be seen, and to create a wireframe of the building;
A second step of displaying the wire frame on a monitor screen of the computer;
A perspective view having two vanishing points left and right virtually read from the memory of the coordinates of the vertices of the reference rectangular parallelepiped set in the virtual spatial coordinate system stored in the memory of the computer and set on the monitor screen A third step of converting the coordinate value into a coordinate value on a coordinate system, and displaying the perspective view of the reference rectangular parallelepiped on the monitor screen in combination with the wire frame of the building;
In response to the operation of the operation unit of the computer, the direction of the reference cuboid is rotated with respect to the spatial coordinate system, and the length in three directions of width, height, and depth is expanded and contracted. A fourth step of moving the positions of the two vanishing points on both the left and right sides to match each ridge line of the reference rectangular parallelepiped to the rectangular parallelepiped portion of the wire frame of the building displayed on the monitor screen;
The coordinate values of the vertices of the reference cuboid with respect to the spatial coordinate system when the rectangular parallelepiped portion of the wireframe of the building and the ridge lines of the reference cube coincide with each other on the monitor screen are stored in the memory. To determine the position of the rectangular parallelepiped portion of the building in the spatial coordinate system,
In response to the operation of the operation unit, the roof of the wire frame of the building displayed on the monitor screen by rotating the upper surface around one of the ridgelines around the upper surface of the reference cuboid A sixth step of aligning the upper surface of the reference cuboid with a portion; and
On the monitor screen, an inclination angle of the upper surface of the reference cuboid with respect to the spatial coordinate system when the upper surface of the reference cuboid coincides with the roof portion of the wire frame of the building is stored in the memory, and the space is determined by the inclination angle. And a seventh step of determining an inclination angle of the roof of the building in the coordinate system.

Images